Ling-flash-2.0 Community
Where it runs 8K context · ≥ 5 t/s floor
| Setup | Best plan | Memory | Generate |
|---|---|---|---|
| 8 GB GPU + 32 GB RAM | Does not fit | ||
| 12 GB GPU + 32 GB RAM | Does not fit | ||
| 16 GB GPU + 64 GB RAM | Q4_K_Mllama.cpp CUDAGPU + CPU experts · 25 expert layers on CPU | Fits | ~4.9–7.3 t/s Estimated extrapolated |
| 24 GB GPU + 64 GB RAM | Q4_K_Mllama.cpp CUDAGPU + CPU experts · 20 expert layers on CPU | Fits | ~6.1–9.0 t/s Estimated extrapolated |
| 32 GB GPU + 128 GB RAM | Q4_K_Mllama.cpp CUDAGPU + CPU experts · 16 expert layers on CPU | Fits | ~7.7–11 t/s Estimated extrapolated |
| 2× 24 GB GPUs + 128 GB RAM | Q4_K_Mllama.cpp CUDAGPU + CPU experts · 8 expert layers on CPU | Fits | ~14–23 t/s Estimated extrapolated |
| 96 GB workstation GPU + 256 GB RAM | Q5_K_Mllama.cpp CUDAFull GPU | Fits | ~73–210 t/s Estimated extrapolated |
| Mac, 24 GB | Does not fit | ||
| Mac, 48 GB | Does not fit | ||
| Mac, 128 GB | Q5_K_Mllama.cpp MetalUnified memory | Fits | ~22–64 t/s Estimated extrapolated |
| Mac, 512 GB | Q8_0llama.cpp MetalUnified memory | Fits | ~23–66 t/s Estimated extrapolated |
| DGX Spark 128 GB | Q5_K_Mllama.cpp CUDAUnified memory | Fits | ~18–48 t/s Estimated |
| Ryzen AI Max+ 395, 128 GB | Q5_K_Mllama.cpp VulkanUnified memory | Fits | ~10–30 t/s Estimated extrapolated |
| CPU only, 64 GB DDR5-5600 dual channel | Does not fit | ||
Same plan engine as the planner, with its default reserves and RAM. How plans are chosen.
Files exact Hugging Face bytes
| Artifact | Size | Files | Revision | Source |
|---|---|---|---|---|
| Q8_0 bartowski/inclusionAI_Ling-flash-2.0-GGUF | 102 GiB | 3 | 1ee418c1a2 | HF files |
| Q5_K_M bartowski/inclusionAI_Ling-flash-2.0-GGUF | 68.28 GiB | 2 | 1ee418c1a2 | HF files |
| Q4_K_M bartowski/inclusionAI_Ling-flash-2.0-GGUF | 58.32 GiB | 2 | 1ee418c1a2 | HF files |
| Native safetensors mixed/custom native; ; stored tensor dtypes=BF16/F32 · BF16/ inclusionAI/Ling-flash-2.0 | 192 GiB | 22 | 96f9b7a73e | HF files |
Not found in the representative quant search (not proven unavailable): GGUF FP16, MLX 8-bit, MLX 4-bit, EXL3, AWQ, GPTQ. Complete selected file sets; projectors, MTP/draft and lookup files are listed separately and budgeted only by plans that use them. File bytes are not allocated memory.
Quality
No Artificial Analysis evaluation mapped to this checkpoint.
Arena text rating
1,361
CI 1354–1368 · 6,947 votes · published 2026-10-15 · dataset
Explicit name identity; exact hosted revision/local quant equivalence not verified
Publisher-reported benchmarks
None extracted from the model card.
AA and Arena describe the hosted model in the stated mode, not each quantized file; scores are never inherited by base, pruned, abliterated or quantized variants and never averaged together.
How to run it
No model-specific recipe in the research. The planner binds the backend’s documented command template (llama.cpp, Ollama, LM Studio, MLX, vLLM) to your plan: open a plan’s details.
Recipes are source examples (settings are the publisher’s or author’s, not measured optima) and were not executed by LocalFit. A downloadable file does not prove your backend build supports the architecture.
Reported speeds 0
No reported measurement yet. Submit one.